The sequence of some regular proteins, when correlated with other structural information, such as data from x-ray diffraction, fiber diffraction, electron microscopy, and spectroscopic analysis, can be used to evaluate models of protein or polymer structure. Three current studies involve the sequence analysis of keratin and other intermediate filaments (with NIDDK, NCI); sequence analysis of streptococcal proteins (with Rockefeller University); and computer models of branched polymers (with PSL, DCRT).